tiny-mlm-glue-qqp-from-scratch-custom-tokenizer-target-glue-stsb

This model is a fine-tuned version of muhtasham/tiny-mlm-glue-qqp-from-scratch-custom-tokenizer on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.0980
• Pearson: 0.1588
• Spearmanr: 0.1567

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Pearson	Spearmanr
3.5246	2.78	500	2.3493	0.0822	0.0866
2.0073	5.56	1000	2.4411	0.1063	0.1054
1.8413	8.33	1500	2.5163	0.1452	0.1350
1.5599	11.11	2000	2.6237	0.1714	0.1662
1.2789	13.89	2500	2.7378	0.1804	0.1772
1.0693	16.67	3000	3.0900	0.1475	0.1275
0.9412	19.44	3500	3.0802	0.1542	0.1499
0.8365	22.22	4000	3.0980	0.1588	0.1567

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu116
• Datasets 2.8.1.dev0
• Tokenizers 0.13.2